tuning nanostructured surfaces with hybrid wettability...
TRANSCRIPT
Nanoscale
PAPER
Cite this: DOI: 10.1039/c8nr05772a
Received 17th July 2018,Accepted 28th September 2018
DOI: 10.1039/c8nr05772a
rsc.li/nanoscale
Tuning nanostructured surfaces with hybridwettability areas to enhance condensation
Shan Gao, Wei Liu* and Zhichun Liu*
Vapor condensation is widespread in natural and industrial applications. Rapid and efficient condensation
plays an essential role in improving energy efficiency. Despite numerous efforts over the past few
decades, the fundamental mechanism of condensation and the microscopic features of condensed dro-
plets are not well understood. Moreover, designing a nanostructured surface with wetting contrast to
enhance dropwise condensation remains unclear. Herein, through molecular dynamics simulation, we
characterized the condensation processes on various nanopillar surfaces, including the nucleation,
growth and coalescence of nanodroplets. During condensation, the droplet size grows linearly with time
as V ∝ t, and the coalescence between small droplets can affect the resultant wetting mode of large dro-
plets. The results indicate that the cooperation between spatially ordering nucleation and dropwise
growth endows hybrid nanopillar surfaces with better heat and mass transfer performance compared
with other homogeneous nanopillar surfaces. Moreover, an interesting dewetting transition occurring on
hydrophobic nanopillar surface was observed during droplet growth, the nucleation site and dewetting
transition were analyzed based on potential energy field of surface. By varying the geometric parameters
of the nanopillar, we found that the condensation rate of the hybrid nanopillar surface increases with the
increase of surface solid fraction. The dense nanopillar array can not only restrain the formation of Wenzel
mode droplet, but also enhance the condensation rate, which provides a guidance for the design of
hybrid nanostructured surfaces.
Introduction
Vapor condensation can be found in a wide range of naturaland industrial applications, including dew formation,1 waterharvesting,2,3 power generation,4 seawater desalination5 andthermal management.6 Among these applications, the efficientcondensation process, as a desired approach to improveenergy efficiency, is dictated by the rapid droplet nucleationand growth rate and the effective removal of condensed dro-plets. Both the geometrical structure and the chemical pro-perties of the condenser surface significantly influence thecondensation.7–10 On a hydrophilic surface, because of thepoor mobility, the accumulative droplets converge to form aliquid film, whose high thermal resistance reduces heat trans-fer. This phenomenon is termed as “filmwise condensation”.For dropwise condensation that occurs on hydrophobic sur-faces, the refresh effect induced by the frequent shedding ofdroplets effectively shortens the existence period of the con-densate. This results in a better heat transfer performance
than filmwise condensation, with an order of magnitudeenhancement.11
Recent advances in the field of micro- and nano-fabricationtechniques facilitate the design and production of superhydro-phobic surfaces,12–15 whose characteristic properties, such asultra-hydrophobicity (contact angle exceed 150°) and extremelylow surface adhesion (rolling angle is less than 10°), makeitself an active research area.16–32 In 2017, through large-scalemolecular dynamics simulations, Zhu et al.33 showed that thetransition between the Cassie and Wenzel states can be con-trolled via precisely designed trapezoidal nanostructures on asurface. It has been lately demonstrated that merged micro-scale droplets can depart from superhydrophobic surfacesindependent of gravity.34 Compared with conventional drop-wise condensation on unstructured hydrophobic surfaces,this distinctive self-removal motion further enhances heattransfer due to the increase in number density of small dro-plets and the decrease in thermal resistance between thesurface and saturated vapor. As such, the droplet-jumpingcondensation on superhydrophobic surfaces has beenextensively investigated in different aspects, including themechanism analysis,8,35–39 the theoretical models of coalesc-ence18,20,40,41 and the design of hierarchically structuredsurfaces.9,19,27–29,33–48 Despite the considerable amount of
School of Energy and Power Engineering, Huazhong University of Science and
Technology (HUST), Wuhan 430074, P. R. China. E-mail: [email protected],
This journal is © The Royal Society of Chemistry 2018 Nanoscale
Publ
ishe
d on
03
Oct
ober
201
8. D
ownl
oade
d by
Hua
zhon
g U
nive
rsity
of
Scie
nce
& T
echn
olog
y on
11/
28/2
018
1:02
:04
AM
.
View Article OnlineView Journal
studies performed, the robustness of superhydrophobic sur-faces in droplet-jumping condensation remains extremely chal-lenging. As schematically shown in Fig. 1, random nucleationat large surface subcooling (ΔT > 5 K) induces the transitionfrom mobile Cassie state to sticky Wenzel state, which greatlyhinders the droplet departure and further leads to a floodingcondensation mode with low heat transfer efficiency.Therefore, an excellent control of Cassie mode nucleation iscrucial to the thermal performance of droplet-jumping con-densation, and it is worthwhile to devote much effort onthis strategy. Inspired by the Namib desert beetle, some hybridsurfaces with heterogeneous wettability49,50 were recently pro-posed to control droplet nucleation and promote the Cassiemode nucleus. However, due to the limited spatial and timeresolution of optical microscopes, the micromorphologicchange of nanodroplet growth in the nucleation stage, whichis essential to the mechanism analysis of droplet nucleationand wetting behavior during condensation, is challenging toexperimentally observe and lacks sufficient understanding.Additionally, to date, it remains unclear how to devise a func-tionalized hybrid surface to achieve the best possible conden-sation heat transfer.
In this study, the molecular dynamics simulation wasadopted to study the condensation processes on differentnanopillar surfaces. The calculated results of droplet growthrate and condensation rate reveal that the hybrid nanopillarsurface has better mass transfer performance compared withother homogeneous nanopillar surfaces. In addition, we ana-lyzed the nucleation site and dewetting transition based onpotential energy field of surface. Furthermore, we found thatthe condensation rate of the hybrid nanopillar surface increaseswith the increase of surface solid fraction. Appropriately redu-cing the structure gap can not only restrain the formation ofWenzel mode droplet, but also enhance the condensation rateof the hybrid nanopillar surface. In the present study, we investi-gated the mechanism of condensation and the microscopic be-
havior of condensed nanodroplets on textured hybrid surfaces.The results provided guidance for the design of functionalizedhybrid surfaces used to enhance condensation.
Methods
All simulations were performed using molecular dynamicssimulation to study the condensation process and evolution ofcondensed nanodroplets on a series of nanopillar surfaceswith different wettability and solid fractions. The simulationdomain is schematically shown in Fig. 2a. Water moleculeswere used as phase change medium. To improve the compu-tation efficiency, simpler copper-like atoms were used to con-stitute the real surfaces with different wettability. To continu-ously provide water vapor as in macroscale condensationexperiments, a hot smooth wall was placed on the upperboundary to heat liquid water. The cold nanostructuredsurface, which comprised a square pillar array with height H =18.1 Å, width W = 23.5 Å and interpillar spacing S = 12.5 Å, wasplaced at the bottom of the box and its horizontal area corres-ponds to 110.5 Å × 110.5 Å. Apart from the hydrophobic nano-pillar surface (Fig. 2c) and hydrophilic nanopillar surface(Fig. 2d), a hybrid surface was established herein. As illustratedin Fig. 2b, the bottom substrate and the side walls of thepillars are hydrophobic, whereas the top surfaces of the pillarsare hydrophilic.
The implementation methods of these models are same asthose in our previous studies.39,51 The periodic boundary con-ditions were only applied in the horizontal direction, while thefixed boundary condition was applied in the lower and upperboundaries. The popular velocity Verlet algorithm with a timestep of 1.0 fs was used to integrate the Newton’s equation ofmotion. The long-range coulombic force was computed by thePPPM (particle–particle particle–mesh) approach. For improv-ing the calculating efficiency, the SHAKE algorithm was usedto fix the bond distance and angle of water molecule. Gravitywas not taken into consideration since the droplets were muchsmaller than the capillary length.
Fig. 1 Schematics of the condensation process on nanostructured sur-faces. (a) Homogeneous superhydrophobic surface. Random nucleationcauses Wenzel mode droplets, whose maximal contact line pingingextremely restrains its mobility. (b) Hybrid surface with heterogeneouswettability. Selective nucleation facilitates Cassie mode droplets, whichcan be easily removed from surfaces via coalescence or gravity.
Fig. 2 (a) Schematic diagram for the simulation domain. Three types ofnanopillar surfaces are constructed here: (b) hydrophobic–hydrophilichybrid surface, (c) hydrophobic surface, (d) hydrophilic surface.
Paper Nanoscale
Nanoscale This journal is © The Royal Society of Chemistry 2018
Publ
ishe
d on
03
Oct
ober
201
8. D
ownl
oade
d by
Hua
zhon
g U
nive
rsity
of
Scie
nce
& T
echn
olog
y on
11/
28/2
018
1:02
:04
AM
. View Article Online
A large-scale atomic/molecular massively parallel simulator(LAMMPS) package was used to conduct all simulations. TheTIP4P model, whose detailed properties were presented in ourprevious studies,45 was employed for water molecules, and theinteraction energy well-depth between water molecules εw wasequal to 0.6487 kJ mol−1. The intermolecular interactionsamong cooper-like atoms and those between cooper-like atomsand water molecules were all modeled by the 12–6 LJ potential.By varying the energy parameter εsw, we adjusted the inter-action intensity between water molecules and cooper-likeatoms. Generally, the lower ratio of energy parameter εsw to εwmakes water molecules less likely to be attracted to thesurface, which indicates that the surface is getting increasinglyhydrophobic. In order to generate hydrophobic surfaces andhydrophilic surfaces, the ratio εsw/εw was set to 0.8 and 1.5,respectively. The snapshots shown in Fig. 3 illustrate the mor-phology of droplets and the corresponding density contourson smooth hydrophobic surface and smooth hydrophilicsurface. The van der Waals and coulombic interactions weretruncated at 8 Å and 12 Å, respectively. In this study, the simu-lation details are almost identical to those of our previousstudies.39,51 The Nose–Hoover thermostat in an NVT ensemblewas applied to heat water molecules and substrate atoms. Afterthe system reaches the desired state, namely, the temperatureof the heated wall, liquid water and cold surface respectivelystabilize at 333 K, 333 K and 285 K, we removed the Nose–Hoover thermostat of water molecules and then, all the watermolecules were integrated with the NVE ensemble.
Results and discussionCondensation on various nanopillar surfaces
First, condensation processes were implemented on hybrid,hydrophobic and hydrophilic nanopillar surfaces. Someselected time-lapse snapshots are shown in Fig. 4; only part ofthe vapor phase is demonstrated in these figures to save space.For hybrid surface, vapor molecules tend to aggregate on thetops of the nanopillars to generate regularly distributed nuclei,which suspend on the surface structure and promote the for-mation of large mobile nanodroplets, i.e., Cassie mode nano-droplets. In contrast, on the hydrophobic surface and hydro-philic surface, the random nucleation behavior induces thepenetration of liquids into the surface structure, which furthergenerates sticky Wenzel mode nanodroplets and liquid film,
respectively. The aforementioned results reveal the inconsis-tency of the nucleus site among various surfaces. A specificanalysis from the point of the potential energy is provided inthe following pages. Additionally, it is interesting to point outthat hybrid nanopillar surface combines the advantages of theother two homogeneous nanopillar surfaces: spatially orderednucleation cooperates with dropwise condensation to generatedesired Cassie droplets, which present superior mobility onnanostructure surfaces.
In order to provide a detailed description of the behaviourof vapor molecules during condensation, simulation results ofthe hybrid surface were selected and visualized, as shown inFig. 5a. In the initial stage, there are random collisionsbetween vapor molecules and the surface. The velocities of thevapor molecules may be reduced after they impact the surfaceatoms, and the decreased part of the kinetic energy is trans-formed into heat, which is absorbed by the surface.Additionally, some other vapor molecules may return into thegas region after collision. Subsequently, numerous vapor mole-cules aggregate into clusters through the van der Waals inter-actions and coulombic force. For the hydrophilic tops of nano-pillars, the attraction of surface atoms to water molecules iscomparably strong. Therefore, the initial clusters tend todeposit on these hydrophilic tops to form regularly distributednuclei, and the released potential energy also translates intoheat. As condensation continues, the nuclei grow up by theaddition of vapor molecules. When the nuclei grow to a sizethat is comparable to the scale of nanopillar tops, they merge
Fig. 4 Condensation processes and the resultant wetting modes onvarious nanopillar surfaces; the upper row of each subfigure representsfront view, and the nether row represents vertical view. (a) Hybrid nano-pillar surface with hydrophilic tops and hydrophobic bases. Hydrophilictops capture vapor molecules to generate spatially ordering nuclei, andhydrophobic bases ensure the dropwise nucleation and growth. Theintegrated effects facilitate the formation of large Cassie droplets. (b)Hydrophobic nanopillar surface. The random nuclei formed on structuretops or in structure gaps induce liquid immerse into structure, yieldingsticky Wenzel droplets. (c) Hydrophilic nanopillar surface. Similarly, therandom nucleation behavior and strong wettability induce the formationof liquid film.
Fig. 3 The equilibrium shape and corresponding density contours ofnanodroplets. (a) On smooth hydrophobic surface. (b) On smoothhydrophilic surface.
Nanoscale Paper
This journal is © The Royal Society of Chemistry 2018 Nanoscale
Publ
ishe
d on
03
Oct
ober
201
8. D
ownl
oade
d by
Hua
zhon
g U
nive
rsity
of
Scie
nce
& T
echn
olog
y on
11/
28/2
018
1:02
:04
AM
. View Article Online
with adjacent nuclei and grow up into droplets.Afterwards, vapor molecules gather on the newly exposedsurface to form new nucleus, and the former positions of dro-plets are migrated after coalescence, which demonstrates theexcellent mobility of condensed droplets. The findingsindicate that during condensation, vapor molecules aggregateinto clusters and deposit on surface to form nuclei, whichgrow up into large Cassie droplets by continuous new nuclea-tion, growth and coalescence. Additionally, it is worth to notethe effects of coalescence on wetting mode transition duringcondensation, as shown in Fig. 5b. Initially, there are fivesmall droplets. Droplets marked as 1, 2 and 3 sit on the hydro-philic tops of nanopillars, while droplets marked as 4 and 5
are constrained in the rough structures. However, the coalesc-ence between droplets 1, 2 and 5 gradually pull the bottomwater molecules out of the valley of pillars, forming a newdroplet 1′. Subsequently, droplet 1′ continues to merge withdroplets 3 and 4 and finally develops into a suspended largedroplet.
As mentioned above, vapor molecules aggregate into clus-ters and deposit on surface to form nuclei, as schematicallydemonstrated in Fig. 6a. Thus, exploring the variation laws ofclusters and nuclei play a critical role in understanding themicroscopic condensation characteristics and nucleationmechanism. Here, we calculated the cluster size andnumber of condensed molecules through a cluster analysisprogram, which extracts related information from atom coordi-nates. A cluster is defined as a group of atoms, where eachatom is within the truncation distance of one or more otheratoms. If the distance between any water molecules in thecluster and the substrate atoms is less than a critical distance(3 Å), the cluster will be identified as a condensed droplet,whose number of molecules can be calculated to evaluate themass transfer performance of surface. The temporal evolutionsof cluster number for all the nanopillar surfaces are shown inFig. 6b. The results indicate that the variation of clusternumber demonstrates a similar tendency to peak function. Inthe initial period of condensation, vapor molecules continuallyaggregate into clusters under the mutual attraction force,resulting in a rapidly increased cluster number. Subsequently,due to the deposition and coalescence of clusters, its numbergradually decreases to a stable value. Fig. 6c shows the evol-ution of maximum nucleus size (the number of water mole-cules contained) on different nanopillar surfaces. A similarvariation trend is observed in each curve: the approximatelylinear increase corresponds to the growth of nucleus byabsorbing vapor molecules, while the abrupt increase resultsfrom the coalescence with other nucleus. In growth stages, dueto sufficient supplement of vapor molecules from the evapor-ation region, the slope of each curve, which represents the
Fig. 5 (a) Time-lapse images of the condensation process on hybridnanopillar surface (orthogonal view). Vapor molecules aggregate intoclusters and deposit on pillar tops to form regularly distributed nuclei,which grow up into large Cassie droplets by continuous new nucleation,growth and coalescence. (b) Coalescence-triggered wetting mode tran-sition of condensed nanodroplets on hybrid nanopillar surface.
Fig. 6 (a) Schematics of the nucleation process: vapor molecules aggregate into clusters and deposit on solid surfaces to form nucleus. (b)Temporal evolution of the number of clusters for different nanopillar surfaces. The points represent calculated values and the fitting curves aremarked red. (c) Temporal evolution of maximum nucleus size for different nanopillar surfaces, the inset shows the dynamic behavior of two dropletsduring coalescence.
Paper Nanoscale
Nanoscale This journal is © The Royal Society of Chemistry 2018
Publ
ishe
d on
03
Oct
ober
201
8. D
ownl
oade
d by
Hua
zhon
g U
nive
rsity
of
Scie
nce
& T
echn
olog
y on
11/
28/2
018
1:02
:04
AM
. View Article Online
growth rate of condensed droplet, remains roughly constant,and the calculated values of droplet growth rate have the fol-lowing relation: hybrid nanopillar surface > hydrophilic nano-pillar surface > hydrophobic nanopillar surface. Additionally,the approximately linear increase in droplet size over timesuggests that the volume of droplet follows the scaling law V ∝t, which is consistent with the experimental data and theore-tical model reported by Hou et al.49
Aforementioned results and analyses reveal that droplets onhybrid nanopillar surface have the highest growth rate. Inorder to further evaluate the mass transfer performance of thesurface, we calculated and recorded the total number of mole-cules deposited on surfaces, whose temporal evolution isshown in Fig. 7a. For all the nanopillar surfaces, there is anincrease in number of condensed molecules. It should benoted that the condensation rate, which is demonstrated bythe slope of each curve, also increases with time and finallyreaches a stable value. It is obvious that the condensation rateof each surface follows the relation hybrid nanopillar surface >hydrophilic nanopillar surface > hydrophobic nanopillarsurface, which indicates the best mass transfer performance ofhybrid nanopillar surface. To provide a quantitative descrip-tion for the wetting state of condensed droplets, we calculatedand recorded the number of water molecules in the structuregap. As shown in Fig. 7b, the variation tendency is consistentwith the observed results. On the hybrid surface, the small andstable amount of molecules in structure gap verifies that thedroplets are in suspended Cassie state. For the hydrophobicsurface, the molecules’ number gradually increases and tendsto a stable value, which corresponds to the growth process ofimmersed Wenzel droplet. For hydrophilic surface, theincrease in number of molecules is the consequence of waterfilm growth and extension. Finally, in the subsequent conden-
sation process, we found an interesting phenomenon, which isshown in Fig. 7c. As the nanodroplet unceasingly grows, adewetting transition occurs when the restriction of nanopillarsto droplet is ineffective: the droplet transforms from immersedWenzel state to suspended Cassie state along with a drasticincrease in contact angle.
Furthermore, the mechanism of condensation nucleus’position distribution and the dewetting transition were ana-lyzed from the perspective of potential energy. According tothe 12–6 LJ potential, we calculated the potential energybetween one water molecule and all surface atoms, and acentral section of elementary unit was selected to plot the con-tours of potential energy, which is shown in Fig. 7d. On thewhole, the value of potential energy in vapor space is 0, andnanopillar surfaces possess two low-potential regions: theregion on the pillar top and the region near the pillar base.Condensation is a process of molecular movement from vaporspace with zero potential energy to surface with negativepotential energy. For the hybrid surface, the pillar top regionhas relatively lower potential energy and is closer to vaporspace. Hence, vapor molecules prefer to aggregate on the topof nanopillars to form spatially ordering nucleus and finallydevelop into a suspended Cassie droplet on the surface struc-ture. For the hydrophobic surface, although the pillar topmore easily captures vapor molecules, the pillar base regionhas lower potential energy. As a result, vapor molecules ran-domly nucleate on the pillar top or in the pillar base andgradually form an immersed Wenzel droplet by coalescence. Atthe start, the droplet is stuck in the nanopillar because its sizeis smaller than the spacing between the pillars. Subsequently,the droplet begins to interact with the surrounding pillarswhen it fills the structure gap. With the further growth ofdroplet, the new condensed molecules deposit on the top of
Fig. 7 (a) Time evolution of the number of condensed molecules (molecules deposited on surface) for different nanopillar surfaces. The slope ofeach curve represents the condensation rate of corresponding surface. (b) Time evolution of the number of molecules in structure gap for differentnanopillar surface. (c) Dewetting transition on hydrophobic nanopillar surface: as the droplet grows, it transfers from immersed Wenzel state to sus-pended Cassie state. (d) Equipotential curves of potential energy between water molecule and different surfaces.
Nanoscale Paper
This journal is © The Royal Society of Chemistry 2018 Nanoscale
Publ
ishe
d on
03
Oct
ober
201
8. D
ownl
oade
d by
Hua
zhon
g U
nive
rsity
of
Scie
nce
& T
echn
olog
y on
11/
28/2
018
1:02
:04
AM
. View Article Online
the droplet, and the restriction effect of surrounding nano-pillars on the droplet gradually fades. The increase in thenumber of molecules in the top region will gradually attractthe bottom molecules out of the structure, resulting in adewetting transition from immersed Wenzel state to sus-pended Cassie state. It is worth noting that the contact angleof the droplet drastically increases after dewetting transition.This is because the contact line of Wenzel droplet is pinned atthe edge of pillars, while the suspended state of Cassie dropletensures a minimal contact line pinning. For the hydrophilicnanopillar surface, similar random nucleation is observed inthe beginning, but the nuclei gradually merge into a waterfilm due to the lower potential energy of surface region (stron-ger attraction of surface atoms).
Condensation on hybrid nanopillar surfaces with differentsolid fractions
Aiming at enhancing condensation, the guidance for design-ing hybrid nanopillar surface is provided herein. By varyingthe width W and interpillar spacing S of the nanopillar, thecondensation processes on four hybrid nanopillar surfaceswith different solid fractions ϕs (ϕs = W2/(W + S)2) were simu-lated and analyzed. Fig. 8a shows the evolution of maximumdroplet size on hybrid nanopillar surfaces with different solidfractions. Similarly, the droplet size linearly increases withtime during the growth stage and rises sharply after coalesc-ence. Although the growth rate of each condensed droplet (theslope of linear portion of each curve) has approximately thesame value, the maximum droplet size increases with theincrease in solid fractions. As shown in Fig. 8b, we calculatedand recorded the total number of molecules deposited on sur-
faces. For all the hybrid nanopillar surfaces with differentsolid fractions, there is an increasing number of condensedmolecules. Clearly, the condensation rate of hybrid nanopillarsurface decreases with the decrease of surface solid fraction.Because the hydrophilic tops easily capture vapor molecules tonucleate, its decreasing proportion weakens the nucleationcapability of hybrid nanopillar surface. It should be noted thatthe wetting modes of the resultant droplet are different oneach surface and present a wetting transition from Cassie stateto Partial wetting state with the decrease of surface solid frac-tion, as shown in Fig. 8c. On the surface with wider structurespacing, the bottom of the droplet is observed to penetrateinto the structure more deeply. Finally, in order to provide ana-lyses for the wetting transition that occurred on hybrid nano-pillar surface, a central section of elementary unit was selectedto plot the contours of potential energy between one watermolecule and all surface atoms. As shown in Fig. 8d, all thehybrid nanopillar surfaces possess two low-potential regions:the region on the pillar top and the region near the pillar base.The pillar top region has relatively lower potential energy. Onsurfaces with high solid fraction, vapor molecules prefer toaggregate on the top of nanopillars to form Cassie droplet. Asinterpillar spacing increases, vapor molecules will contact withthe pillar base region more frequently and condense in struc-ture gaps with a higher probability, inducing the formation ofimmersed Wenzel droplet. However, the Cassie droplet is mostdesirable due to its excellent mobility. Hence, the crucialdesign principle of nanostructured surface with highlyefficient dropwise condensation heat transfer is to promotethe nucleation and growth of droplet on the structure top.Based on the above results, we conclude that properly reducing
Fig. 8 (a) Temporal evolution of maximum nucleus size for hybrid nanopillar surfaces with different solid fractions. (b) Time evolution of thenumber of condensed molecules (molecules deposited on surface) for hybrid nanopillar surfaces with different solid fractions. The slope of eachcurve represents the condensation rate of corresponding surface. (c) The wetting mode of resultant droplet on each hybrid nanopillar surface: as thesolid fraction decreases, it transfers from suspended Cassie state to partial wetting state. (d) Equipotential curves of potential energy between watermolecule and hybrid nanopillar surfaces with different solid fractions.
Paper Nanoscale
Nanoscale This journal is © The Royal Society of Chemistry 2018
Publ
ishe
d on
03
Oct
ober
201
8. D
ownl
oade
d by
Hua
zhon
g U
nive
rsity
of
Scie
nce
& T
echn
olog
y on
11/
28/2
018
1:02
:04
AM
. View Article Online
the structure gap can not only restrain the formation ofWenzel mode droplet, but also enhance the condensation rateof hybrid nanopillar surface.
Conclusions
In summary, the condensation processes on different nano-pillar surfaces were studied numerically using the moleculardynamics simulation. Compared with homogeneous nano-pillar surfaces, the regular nucleation and dropwise conden-sation facilitates the formation of Cassie droplets and furtherendows the hybrid nanopillar surfaces with better mass trans-fer performance. During condensation, the coalescencebetween small droplets can influence the wetting mode ofresultant large droplets and an interesting dewetting transitionoccurs on hydrophobic nanopillar surface. By varying the geo-metric parameters of pillar for hybrid nanopillar surfaces, itwas found that the surface solid fraction plays a crucial role inthe condensation enhancement. As the solid fractionincreases, the condensation rate of hybrid nanopillar surfaceincreases and the wetting mode of resultant droplet transfersfrom partial wetting state to the suspended Cassie state.Hence, properly reducing the structure gap can not onlyrestrain the formation of Wenzel mode droplet, but alsoenhance the condensation rate of hybrid nanopillar surface.We believe that these findings can provide a microscopic andfundamental insight into the mechanism of condensation andare conducive to the design of hybrid nanostructure surfacesused to enhance heat and mass transfer during dropwisecondensation.
Conflicts of interest
The authors declare no competing financial interest.
Acknowledgements
This project was supported by the National Natural ScienceFoundation of China (No. 51736004 and No. 51776079), andthe National Key Research and Development Program of China(No. 2017YFB0603501-3). The study was performed at theNational Supercomputer Center in Tianjin, and the calcu-lations were performed on TianHe-1(A).
References
1 N. Agam and P. R. Berliner, J. Arid Environ., 2006, 65, 572.2 X. Dai, N. Sun, S. O. Nielsen, B. B. Stogin, J. Wang, S. Yang
and T. S. Wong, Sci. Adv., 2018, 4, eaaq0919.3 J. Ju, H. Bai, Y. Zheng, T. Zhao, R. Fang and L. Jiang, Nat.
Commun., 2012, 3, 1247.4 J. Lee, S. Mahendra and P. J. J. Alvarez, ACS Nano, 2010, 4,
3580.
5 A. D. Khawaji, I. K. Kutubkhanah and J. M. Wie,Desalination, 2008, 221, 47.
6 S. Daniel, M. K. Chaudhury and J. C. Chen, Science, 2001,291, 633.
7 M. D. Mulroe, B. R. Srijanto, S. F. Ahmadi, C. P. Collier andJ. B. Boreyko, ACS Nano, 2017, 11, 8499.
8 R. Enright, N. Miljkovic, A. A. Obeidi, C. V. Thompson andE. N. Wang, Langmuir, 2012, 28, 14424.
9 R. Wang, J. Zhu, K. Meng, H. Wang, T. Deng, X. Gao andL. Jiang, Adv. Funct. Mater., 2018, 29, 1800634.
10 N. Miljkovic, R. Enright, Y. Nam, K. Lopez, N. Dou, J. Sackand E. N. Wang, Nano Lett., 2013, 13, 179.
11 J. W. Rose, Proc. Inst. Mech. Eng., Part A, 2002, 216, 115.12 P. Guo, M. Wen, L. Wang and Y. Zheng, Nanoscale, 2014, 6,
3917.13 K. Koch, B. Bhushan, Y. C. Jung and W. Barthlott, Soft
Matter, 2009, 5, 1386.14 K. H. Chu, R. Xiao and E. N. Wang, Nat. Mater., 2010, 9,
413.15 E. Zhang, Y. Wang, T. Lv, L. Li, Z. Cheng and Y. Liu,
Nanoscale, 2015, 7, 6151.16 H. Cha, J. M. Chun, J. Sotelo and N. Miljkovic, ACS Nano,
2016, 10, 8223.17 J. Oh, R. Zhang, P. P. Shetty, J. A. Krogstad, P. V. Braun and
N. Miljkovic, Adv. Funct. Mater., 2018, 28, 1707000.18 J. Tian, J. Zhu, H. Y. Guo, J. Li, X. Q. Feng and X. Gao,
J. Phys. Chem. Lett., 2014, 5, 2084.19 Y. Zhao, Y. Luo, J. Zhu, J. Li and X. Gao, ACS Appl. Mater.
Interfaces, 2015, 7, 11719.20 K. Wang, Q. Liang, R. Jiang, Y. Zheng, Z. Lan and X. Ma,
RSC Adv., 2016, 6, 99314.21 S. Baek, W. Kim, S. Jeon and K. Yong, Nanoscale, 2017, 9,
6665.22 W. S. Wong, G. Liu, N. Nasiri, C. Hao, Z. Wang and
A. Tricoli, ACS Nano, 2017, 11, 587.23 C. Hao, Y. Liu, X. Chen, J. Li, M. Zhang, Y. Zhao and
Z. Wang, Small, 2016, 12, 1825.24 C. Hao, J. Li, Y. Liu, X. Zhou, Y. Liu, R. Liu, L. Che,
W. Zhou, D. Sun, L. Li, L. Xu and Z. Wang, Nat. Commun.,2015, 6, 7986.
25 Y. Hou, M. Yu, Y. Shang, P. Zhou, R. Song, X. Xu,X. Chen, Z. Wang and S. Yao, Phys. Rev. Lett., 2018, 120,075902.
26 J. Li, X. Zhou, J. li, L. Che, J. Yao, G. McHale,M. K. Chaudhury and Z. Wang, Sci. Adv., 2017, 3, eaao3530.
27 J. Li, R. Kang, X. Tang, H. She, Y. Yang and F. Zha,Nanoscale, 2016, 8, 7638.
28 X. Dai, B. B. Stogin, S. Yang and T. S. Wong, ACS Nano,2015, 9, 9260.
29 R. Wen, Q. Li, W. Wang, B. Latour, C. H. Li, C. Li, Y. C. Leeand R. Yang, Nano Energy, 2017, 38, 59.
30 T. Koishi, K. Yasuoka and X. C. Zeng, Langmuir, 2017, 33,10184.
31 J. Liu, C. Zhu, K. Liu, Y. Jiang, Y. Song, J. S. Francisco,X. C. Zeng and J. Wang, Proc. Natl. Acad. Sci. U. S. A., 2017,114, 11285.
Nanoscale Paper
This journal is © The Royal Society of Chemistry 2018 Nanoscale
Publ
ishe
d on
03
Oct
ober
201
8. D
ownl
oade
d by
Hua
zhon
g U
nive
rsity
of
Scie
nce
& T
echn
olog
y on
11/
28/2
018
1:02
:04
AM
. View Article Online
32 D. Niu and G. Tang, RSC Adv., 2018, 8, 24517.33 C. Zhu, Y. Gao, Y. Huang, H. Li, S. Meng, J. S. Francisco
and X. C. Zeng, Nanoscale, 2017, 9, 18240.34 J. B. Boreyko and C. H. Chen, Phys. Rev. Lett., 2009, 103,
184501.35 S. Chavan, H. Cha, D. Orejon, K. Nawaz, N. Singla,
Y. F. Yeung, D. Park, D. H. Kang, Y. Chang, Y. Takata andN. Miljkovic, Langmuir, 2016, 32, 7774.
36 H. Cha, A. Wu, M. K. Kim, K. Saigusa, A. Liu andN. Miljkovic, Nano Lett., 2017, 17, 7544.
37 X. Chen, R. Ma, H. Zhou, X. Zhou, L. Che, S. Yao andZ. Wang, Sci. Rep., 2013, 3, 2515.
38 Q. Hao, Y. Pang, Y. Zhao, J. Zhang, J. Feng and S. Yao,Langmuir, 2014, 30, 15416.
39 S. Gao, Q. Liao, W. Liu and Z. Liu, J. Phys. Chem. Lett.,2018, 9, 13.
40 K. Rykaczewski, A. T. Paxson, S. Anand, X. Chen, Z. Wangand K. K. Varanasi, Langmuir, 2013, 29, 881.
41 J. Zhu, Y. Luo, J. Tian, J. Li and X. Gao, ACS Appl. Mater.Interfaces, 2015, 7, 10660.
42 S. Gao, Q. Liao, W. Liu and Z. Liu, J. Phys. Chem. C, 2018,122, 20521.
43 H. Li, J. Zhu, Y. Luo, F. Yu, J. Fang and X. Gao, Adv. Mater.Interfaces, 2016, 3, 1600362.
44 Q. Zhang, M. He, J. Chen, J. Wang, Y. Song and L. Jiang,Chem. Commun., 2013, 49, 4516.
45 S. Gao, Q. Liao, W. Liu and Z. Liu, Langmuir, 2017, 33,12379.
46 R. Wen, S. Xu, X. Ma, Y. C. Lee and R. Yang, Joule, 2018, 2,269.
47 R. Wen, Q. Li, J. Wu, G. Wu, W. Wang, Y. Chen, X. Ma,D. Zhao and R. Yang, Nano Energy, 2017, 33, 177.
48 R. Wen, Z. Lan, B. Peng, W. Xu, R. Yang and X. Ma, ACSAppl. Mater. Interfaces, 2017, 9, 13770.
49 Y. Hou, M. Yu, X. Chen, Z. Wang and S. Yao, ACS Nano,2015, 9, 71.
50 X. Dai, X. Huang, F. Yang, X. Li, J. Sightler, Y. Yang andC. Li, Appl. Phys. Lett., 2013, 102, 161605.
51 S. Gao, Q. Liao, W. Liu and Z. Liu, Langmuir, 2018, 34,5910.
Paper Nanoscale
Nanoscale This journal is © The Royal Society of Chemistry 2018
Publ
ishe
d on
03
Oct
ober
201
8. D
ownl
oade
d by
Hua
zhon
g U
nive
rsity
of
Scie
nce
& T
echn
olog
y on
11/
28/2
018
1:02
:04
AM
. View Article Online